Alzheimer's disease (AD) is the most frequent neurodegenerative disorder in developed countries, accounting for 60-70% of all dementia cases. Around 5% of all people over the age of 65 years are suffering from AD. Due to a close correlation to age the number of patients is dramatically increasing worldwide. Around 5% of all people above the age of 65 years are suffering from AD while about 30-50% over the age of 80 are affected. In the US alone an estimated 4.5 million patients (in Germany 1.5 million) have Alzheimer's disease. This number is expected to grow to up to 16 million people in 2050. The spending in 2005 was 91 billion US-$ for medication and nursing in the USA. With respect to the increasing life expectancy, the development of a treatment strategy is the present therapeutic challenge and due to a close correlation to age, the number of patients is dramatically increasing worldwide. Even after two decades of intense research, there is currently no treatment available, which rebuilds or even sustains the cognitive function of the patients in a long-lasting manner.
Present endeavors for potential new causal treatments include application of                (i) intercalating substances to prevent aggregation of neurotoxic Aβ peptides,        (ii) the inhibition of the enzymes responsible for Aβ formation, or        (iii) clearance of the Aβ-aggregates by vaccination approaches (reviewed in: Hardy and Selkoe (2002) Science 297, 353-356 and Citron (2004) Trends Pharmacol Sci. 25, 92-97).        
With regard to vaccination, recent results point to dramatic adverse effects: phase II trials failed because of a high percentage of aseptic encephalitis cases (Patton, et al. (2006) Am J Pathol. 169, 1048-63). Also, the apparently straightforward approach to develop β-secretase (BACE-1) and γ-secretase inhibitors is hampered by their importance for basic cellular processes, e.g. is BACE-1 activity essential for myelination of axons (Glabe, C. (2006) Science 314, 602-603). Also, APP and even Aβ seem to fulfill important physiological functions (Pearson and Peers (2006) J Physiol. 575, 5-10), which might obstruct therapeutic strategies to suppress Aβ generation in general.
A characteristic symptom of AD is the progressive cognitive impairment, characterized by loss of memory, function, language abilities, judgment and executive functioning. Later disease stages are often associated with severe behavioral symptoms (aggression, delusions, hallucinations, disturbed day-night cycle) and the loss of activities of daily living. Reasons for the observed cognitive decline are molecular and histopathological changes in the brain. The earliest events are loss of synaptic contacts in the entorhinal cortex. At later stages, there is also a significant loss of cholinergic neurons in basal forebrain nuclei, a finding that led to the cholinergic theory behind memory loss in this disease and which was the basis for the development of a first symptomatic treatment. Pathological hallmarks of the disease are extracellular amyloid deposits consisting mainly of the amyloid beta peptide (neuritic plaques), and intracellular neurofibrillary tangles which are formed by the hyperphosphorylated microtubule-associated protein tau. The cerebral cortex and the hippocampal regions are particularly affected. According to the amyloid hypothesis, initial amyloid aggregates are caused by intracellular Aβ accumulation, which then initiates the pathophysiologic cascade including plaque formation, neuroinflammation and tangle-formation (Selkoe (2001) Physiol. Rev. 81, 741-766).
The core of amyloid plaques in AD consists of multimeric aggregates of a polypeptide of 40 or 42 residues (4 kDa), depending on the cleavage site of γ-secretase, called amyloid beta or Aβ. The peptide is generated from the amyloid precursor protein (APP), a class 1 transmembrane protein which is highly expressed in neuronal cells, by successive proteolysis of a β- and γ-secretase (FIG. 22-1). In contrast, cleavage of APP by α-secretase precludes generation of Aβ. Therefore, α-cleavage is the physiological, non-amyloidogenic counterpart to the processing by the other secretases of the amyloidogenic pathway.
Depending on the chain length, Aβ peptides display different neurotoxicity, i.e. the longer form Aβ(1-42) is particularly important for developing Alzheimer's disease. It has been shown in vitro that Aβ(1-42)-peptides aggregate more rapidly than Aβ(1-40) peptides. Furthermore, analysis of brains derived from patients with sporadic Alzheimer's disease and Down's syndrome, who inevitably develop AD as a result of the presence of a third APP gene, has shown that Aβ(1-42) peptides deposit at the beginning and in a highly selective manner in senile plaques. These findings provided strong evidence that aggregation and deposition of Aβ(1-42) peptides may be a common initiating event in all forms of AD.
So, in the last decades, research was focused on the C-terminal part of the Aβ-peptides and the influence of C-terminal modifications on AD development and progression. More recently, research was started on N-terminal modifications of Aβ(1-40) and Aβ(1-42). Several new studies have shown that the Aβ-peptides deposited in brains of AD patients are heterogeneous at the N-terminus. The N-terminus of Aβ is generated by β-secretase cleavage at position Asp1 (β-cleavage) and to a much lesser extent at Glu11 (β′-cleavage) of Aβ. However, it has been shown, that the peptides in plaques display pronounced heterogeneity in terms of the N-terminal amino acid. The L-aspartate residues, normally present at positions 1 and 7 in β-amyloid peptides, can be isomerized to iso-aspartate or racemized to form D-aspartate. More importantly, the glutamates are cyclized to pyroglutamic acid (pGlu) in truncated peptides starting at position 3 or 11. Analysis of brains from AD or DS patients show that the core of amyloid plaques consists of β-amyloid peptides containing N-terminal pyroglutamate, e.g. pGlu3-Aβ(3-40/42) and pGlu11-Aβ(11-40/42). These shortened and modified peptides can account for about 50% of the whole Aβ deposited in plaques (Harigaya, et al. (2000) BBRC 276, 422-427; Sergeant, et al. (2003) J. Neurochem. 85, 1581-1591; Russo, et al. (1997) FEBS Lett. 409, 411-416; Kuo, et al. (1997), BBRC 237, 188-191). Moreover, they are reported to be more neurotoxic and to aggregate more rapidly than full-length isoforms in vitro (Schilling, et al. (2006), Biochemistry 45, 12393-12399; Russo, et al. (2002), J. Neurochem. 82, 1480-1489; He and Barrow (1999), Biochemistry 38, 10871-10877). In a recent study, normal-aged (NA) elderly persons with plaque deposits, but without showing Alzheimer's pathology, could be distinguished from patients suffering from sporadic Alzheimer's disease by characterization of the composition of Aβ species (Piccini, et al. (2005), J. Biol. Chem., 34186-34192). In water-soluble brain fractions of patients with AD, a significantly higher amount of N-terminally truncated, pGlu-modified Aβ peptides was detected, suggesting that accumulation of these species represents a nidus for development of neurodegeneration. A correlation of the insoluble pGlu-Aβ concentration with the state of disease progression was also reported very recently (Güntert, et al. (2006) Neuroscience, 143, 461-475). Additionally, elevated levels of N-truncated Aβ peptides were identified in familial AD cases caused by presenilin mutations. In fact, this further substantiates the relevance of N-modified Aβ for the severity of the disease (Miravalle, et al. (2005), Biochemistry 44, 10810-10821; Russo, et al. (2000), Nature 405, 531-532). Finally, it is important to consider that pGlu-modified Aβ peptides are resistant to degradation by extracellular aminopeptidases. Therefore, pyroglutamate-modified Aβ species have, a prolonged half-life in vivo, which favors accumulation and formation of neurotoxic aggregates (Saido (1998) Neurobiol. Aging, 19, S69-S75). Taken together, the results indicate that the pGlu3-Aβ peptides play an important, probably even the decisive, role in the development of Alzheimer's disease.
Glutaminyl Cyclases—Catalysts of Pyroglutamyl Formation In Vivo
Several bioactive hormones, e.g. Thyrotropin-Releasing Hormone (TRH) or Gastrin, possess a N-terminal pyroglutamic acid residue, which is generated from glutamine during prohormone maturation. Glutaminyl cyclases (QCs) have been identified to catalyze the cyclization of N-terminal glutaminyl residues in plants and animals under concomitant release of ammonia (Messer, (1963), Nature 4874, 1299; Fischer and Spiess, (1987), PNAS 84, 3628-3632).
A QC was first isolated by Messer from the latex of the tropical plant Carica papaya in 1963 (Messer, M. (1963) Nature 4874, 1299). 24 years later, a corresponding enzymatic activity was discovered in animal pituitary (Busby, W. H. J. et al. (1987) J Biol Chem 262, 8532-8536; Fischer, W. H. and Spiess, J. (1987) Proc Natl Acad Sci USA 84, 3628-3632). For the mammalian QCs, the conversion of Gln into pGlu by QC could be shown for the precursors of TRH and GnRH (Busby, W. H. J. et al. (1987) J Biol Chem 262, 8532-8536; Fischer, W. H. and Spiess, J. (1987) Proc Natl Acad Sci USA 84, 3628-3632). In addition, initial localization experiments of QC revealed a co-localization with its putative products of catalysis in bovine pituitary, further improving the suggested function in peptide hormone synthesis (Bockers, T. M. et al. (1995) J Neuroendocrinol 7, 445-453). In contrast, the physiological function of the plant QC is less clear. In case of the enzyme from C. papaya, a role in the plant defense against pathogenic microorganisms was suggested (El Moussaoui, A. et al. (2001) Cell Mol Life Sci 58, 556-570). Recently, other QCs from plants were identified by sequence comparisons (Dahl, S. W. et al. (2000) Protein Expr Purif 20, 27-36; Schilling, S. et al. (2007) Biol Chem 388, 145-153) The physiological function of these enzymes, is presumably the pGlu-formation at the N-terminus of pathogenesis-related proteins. The QCs known from plants and animals show a strict specificity for L-Glutamine at the N-terminal position of the substrates and their kinetic behavior was found to obey the Michaelis-Menten equation mostly (Pohl, T. et al. (1991) Proc Natl Acad Sci USA 88, 10059-10063; Consalvo, A. P. et al. (1988) Anal Biochem 175, 131-138; Gololobov, M. Y. et al. (1996) Biol Chem Hoppe Seyler 377, 395-398, Schilling, S. et al. (2003) Biol Chem 384, 1583-1592). A comparison of the primary structures of the QC from C. papaya and of the highly conserved QCs from mammals, however, did not reveal any sequence homology (Dahl, S. W. et al. (2000) Protein Expr Purif 20, 27-36). Whereas the plant QCs appear to belong to a new enzyme family (Dahl, S. W. et al. (2000) Protein Expr Purif 20, 27-36), the mammalian QCs were found to have a pronounced sequence homology to bacterial aminopeptidases (Bateman, R. C. et al. (2001) Biochemistry 40, 11246-11250), leading to the conclusion that the QCs from plants and animals have different evolutionary origins.
EP 02 011 349.4 discloses polynucleotides encoding insect glutaminyl cyclase, as well as polypeptides encoded thereby. This application further provides host cells comprising expression vectors comprising polynucleotides of the invention. Isolated polypeptides and host cells comprising insect QC are useful in methods of screening for agents that reduce glutaminyl cyclase activity. Such agents are described as useful pesticides.
First tissue distribution studies of mammalian QCs revealed that they are mainly expressed in brain and some peripheral glands, e.g. thyroid and thymus. The enzyme is expected to be directed to the regulated secretory pathway of the expressing cells where the hormone maturation process takes place (Bockers, et al. (1995), J. Neuroendocrinol. 7, 445-453). Upon stimulation, QCs appear to be secreted from the cells together with the mature hormones.
Mammalian QCs were isolated and characterized from human, bovine and murine sources (Pohl, et al. (1991), PNAS 88, 10059-19963; Schilling, et al. (2002), Biochemistry 41, 10849-10857; Schilling, et al. (2005), Biochemistry 44, 13415-13424). The open reading frames of the enzymes consist of 361 (human, bovine) or 362 (murine) amino acids with an overall sequence identity of about 80%. The mature proteins are glycosylated. Human QC has been shown to be highly specific for the L-configuration of glutamine at the N-terminal amino acid position. Free glutamine was not converted (Schilling, et al. (2003) Biol Chem 384, 1583-1592). Other restrictions, however, were not observed, implying that the enzyme is responsible for N-terminal conversion of the very heterogenous group of pGlu-hormones and proteins (Schilling, et al. (2003), Biol. Chem. 384, 1583-1592). Initial mechanistic studies on human QC pointed to an involvement of histidinyl residues in the binding and conversion of the substrate as indicated by diethyl pyrocarbonate inhibition and site directed mutagenesis (Bateman, et al. (2001), Biochemistry 40, 11246-11250). Later evidence showed that mammalian QCs are structurally related to zinc-dependent aminopeptidases and that the residues for complexation of the active site metal ions are also conserved in human QC. Furthermore, 1,10-phenanthroline, dipicolinic acid and imidazole inhibited human QC, implying a metal-dependent catalysis of the enzyme (Schilling, et al. (2003), J. Biol. Chem. 278, 49773-49779). Determination of metal content of human and murine QC and crystallization revealed a single zinc-dependent catalytic in human and murine QC mechanism (Huang, et al. (2005), PNAS 102, 13117-13122; Schilling, et al. (2005), Biochemistry 44, 13415-13424).
In contrast to the physiological QC-substrates, glutamic acid is the precursor of pGlu at the N-terminus of Aβ. Initially, it was suggested that cyclization of glutamic acid proceeds spontaneously (Hashimoto, et al. (2002), EMBO J. 21, 1524-1534). The identification of QCs in brain regions vulnerable to AD, however, triggered research by the present inventors on the catalytic properties of QC. Indeed, catalysis of pGlu-Aβ(3-11) and pGlu-Aβ(3-21) generation could be demonstrated in vitro (Schilling, et al. (2004), FEBS Lett. 563, 191-196). Interestingly, the pH-dependence of catalysis reveals an optimal substrate conversion under mildly acidic conditions, which contrasts with the basic pH-optimum for cyclization of glutaminyl peptides. This, in fact, supports a conversion of glutamic acid at the N-terminus of Aβ in the secretory pathway, were an acidic pH environment has been described. Pyroglutamate present at the N-terminus of the C-terminal fragments supports such a conclusion (Russo, et al. (2001), Neurobiol. Dis. 8, 173-180). Additionally, an involvement of QC in conversion of glutamic acid into pGlu is supported by the following observations:                1. APP, the precursor of Aβ, and QC are highly abundant in brain tissue,        2. QC and APP are expressed in the secretory pathway,        3. Obviously, Aβ is generated at least partially in secretory compartments. Hence, QC and Aβ (or the C-terminal Fragment of APP) might be colocalized intracellularly at high concentrations of both species, thus promoting conversion. Other peptides and proteins originating from high-level QC expressing tissue are known carrying N-terminal pGlu originating from glutamic acid, e.g. the pituitary-derived hormone β-Lipotropin (β-LPH) or immunoglobulins and the amyloidogenic ADan and ABri peptides as well (Bateman, et al. (1991) J. Biol. Chem. 265, 22130-22136; Twardzik and Peterkovsky (1972) PNAS 69, 274-7, Ghiso, et al. (2001) Amyloid 8, 277-284). (see also FIG. 22-2).        Chemotactic cytokines (chemokines) are proteins that attract and activate leukocytes and are thought to play a fundamental role in inflammation. Chemokines are divided into four groups categorized by the appearance of N-terminal cysteine residues (“C”-; “CC”-; “CXC”- and “CX3C”-chemokines). “CXC”-chemokines preferentially act on neutrophils. In contrast, “CC”-chemokines attract preferentially monocytes to sites of inflammation. Monocyte infiltration is considered to be a key event in a number of disease conditions (Gerard and Rollins (2001) Nat. Immunol 2, 108-115; Bhatia, et al. (2005) Pancreatology. 5, 132-144; Kitamoto, et al. (2003) J Pharmacol Sci. 91, 192-196). The MCP family, as one family of chemokines, consists of four members (MCP-1-4), displaying a preference for attracting monocytes but showing differences in their potential (Luini, et al. (1994) Cytokine 6, 28-31; Uguccioni, et al. (1995) Eur J Immunol 25, 64-68). In the following both cDNA as well as amino acid sequences of MCP-1-4 are indicated:        
Human MCP-1 (CCL2) (GeneBank Accession: M24545)cDNA (300 bp)SEQ ID NO: 4  1 atgaaagtct ctgccgccct tctgtgcctg ctgctcatag cagccacctt cattccccaa  61 gggctcgctc agccagatgc aatcaatgcc ccagtcacct gctgttataa cttcaccaat 121 aggaagatct cagtgcagag gctcgcgagc tatagaagaa tcaccagcag caagtgtccc 181 aaagaagctg tgatcttcaa gaccattgtg gccaaggaga tctgtgctga ccccaagcag 241 aagtgggttc aggattccat ggaccacctg gacaagcaaa cccaaactcc gaagacttga Protein (Signal Sequence in bold: 23 aa; Mature MCP-1: 76 aa)SEQ ID NO: 5MKVSAALLCLLLIAATFIPQGLAQPDAINAPVTCCYNFTNRKISVQRLASYRRITSSKCPKEAVIFKTI VAKEICADPKQKWVQDSMDHLDKQTQTPKT Human MCP-2 (CCL8) (GeneBank Accession: Y10802)cDNA (300 bp)SEQ ID NO: 6  1 atgaaggttt ctgcagcgct tctgtgcctg ctgctcatgg cagccacttt cagccctcag  61 ggacttgctc agccagattc agtttccatt ccaatcacct gctgctttaa cgtgatcaat 121 aggaaaattc ctatccagag gctggagagc tacacaagaa tcaccaacat ccaatgtccc 181 aaggaagctg tgatcttcaa gacccaacgg ggcaaggagg tctgtgctga ccccaaggag 241 agatgggtca gggattccat gaagcatctg gaccaaatat ttcaaaatct gaagccatga Protein (Signal Sequence in bold: 23 aa; Mature MCP-2: 76 aa)SEQ ID NO: 7MKVSAALLCLLLMAATFSPQGLAQPDSVSIPITCCENVINRKIPIQRLESYTRITNIQCPKEAVIFKTQ RGKEVCADPKERWVRDSMKHLDQIFQNLKP Human MCP-3 (CCL7) (GeneBank Accession: X71087)cDNA (300 bp)SEQ ID NO: 8  1 atgaaagcct ctgcagcact tctgtgtctg ctgctcacag cagctgcttt cagcccccag  61 gggcttgctc agccagttgg gattaatact tcaactacct gctgctacag atttatcaat 121 aagaaaatcc ctaagcagag gctggagagc tacagaagga ccaccagtag ccactgtccc 181 cgggaagctg taatcttcaa gaccaaactg gacaaggaga tctgtgctga ccccacacag 241 aagtgggtcc aggactttat gaagcacctg gacaagaaaa cccaaactcc aaagctttga Protein (Signal Sequence in bold: 23 aa; Mature MCP-3: 76 aa)SEQ ID NO: 9MKASAALLCLLLTAAAFSPQGLAQPVGINTSTTCCYRFINKKIPKQRLESYRRTTSSHCPREAVIFKTK LDKEICADPTQKWVQDFMKHLDKKTQTPKL MCP-4 (CCL13) (GeneBank Accession: U46767)cDNA (297 bp)SEQ ID NO: 10  1 atgaaagtct ctgcagtgct tctgtgcctg ctgctcatga cagcagcttt caacccccag  61 ggacttgctc agccagatgc actcaacgtc ccatctactt gctgcttcac atttagcagt 121 aagaagatct ccttgcagag gctgaagagc tatgtgatca ccaccagcag gtgtccccag 181 aaggctgtca tcttcagaac caaactgggc aaggagatct gtgctgaccc aaaggagaag 241 tgggtccaga attatatgaa acacctgggc cggaaagctc acaccctgaa gacttga Protein (Signal Sequence in bold: 23 aa; Mature MCP-4: 75 aa)SEQ ID NO: 11MKVSAVLLCLLLMTAAFNPQGLAQPDALNVPSTCCFTFSSKKISLQRLKSYVITTSRCPQKAVIFRTKL GKEICADPKEKWVQNYMKHLGRKAHTLKT
The inventors have shown that the mature form of human and rodent MCP-1 is posttranslationally modified by Glutaminyl Cyclase (QC), resulting in the formation of an N-terminal pyroglutamyl (pGlu) residue. The N-terminal pGlu modification confers resistance against N-terminal degradation by aminopeptidases, which is of importance, since chemotactic potency of MCP-1 is mediated by its N-terminus (Van Damme, J., et al., (1999) Chem Immunol 72, 42-56). Artificial elongation or degradation leads to a loss of function although MCP-1 still binds to its receptor (CCR2) (Proost, P., et al., (1998), J Immunol 160, 4034-4041; Zhang, Y. J., et al., (1994), J Biol. Chem. 269, 15918-15924; Masure, S., et al., 1995, J Interferon Cytokine Res. 15, 955-963; Hemmerich, S., et al., (1999) Biochemistry 38, 13013-13025).
Due to the major role of MCP-1 in a number of disease conditions, an anti-MCP-1 strategy is urgently needed. Therefore, small orally available compounds inhibiting the action of MCP-1 are promising candidates for a drug development. Inhibitors of Glutaminyl Cyclase are small orally available compounds, which target the important step of pGlu-formation at the N-terminus of MCP-1 (Cynis, H., et al., (2006) Biochim. Biophys. Acta 1764, 1618-1625; Buchholz, M., et al., (2006) J Med Chem 49, 664-677). In consequence, after application of a QC-inhibitor the N-terminus of MCP-1 is not protected by a pGlu-residue possessing a glutamine-proline motif, which can be cleaved by dipeptidylpeptidases, e.g. dipeptidylpeptidase 4 and fibroblast activating protein (FAP, Seprase), abundantly existing on the endothelium and within the blood circulation. This leads to the formation of N-terminal truncated MCP 1 unfolding an antagonistic action at the CCR2 and therefore, inhibiting monocyte-related disease conditions.
As mentioned above, Monocyte chemoattractant protein 1 (MCP-1, CCL2) belongs to a family of potent chemotactic cytokines (CC chemokines), that regulate the trafficking of leukocytes, especially monocytes, macrophages and T-cells, to sites of inflammation (Charo, I. F. and Taubman, M. B. (2004) Circ. Res. 95, 858-866). Besides its role in, e.g. vascular disease, compelling evidence points to a role of MCP 1 in Alzheimer's disease (AD) (Xia, M. Q. and Hyman, B. T. (1999) J Neurovirol. 5, 32-41). The presence of MCP-1 in senile plaques and in reactive microglia, the residential macrophages of the CNS, has been observed in brains of patients suffering from AD (Ishizuka, K., et al., (1997) Psychiatry Clin. Neurosci. 51, 135-138). Stimulation of monocytes and microglia with Amyloid-β protein (Aβ) induces chemokine secretion in vitro (Meda, L., et al., (1996) J Immunol 157, 1213-1218; Szczepanik, A. M., et al., (2001) J Neuroimmunol. 113, 49-62) and intracerebroventricular infusion of Aβ (1-42) into murine hippocampus significantly increases MCP-1 in vivo. Moreover, Aβ deposits attract and activate microglial cells and force them to produce inflammatory mediators such as MCP-1, which in turn leads to a feed back to induce further chemotaxis, activation and tissue damage. At the site of Aβ deposition, activated microglia also phagocyte Aβ peptides leading to activation (Rogers, J. and Lue, L. F. (2001) Neurochem. Int. 39, 333-340).
Examination of chemokine expression in the 3×Tg mouse model for AD revealed that neuronal inflammation precedes plaque formation and MCP-1 is upregulated by a factor of 11. Furthermore, the upregulation of MCP-1 seems to correlate with the occurrence of first intracellular Aβ deposits (Janelsins, M. C., et al., (2005) J Neuroinflammation. 2, 23). Cross-breeding of the Tg2575 mouse model for AD with a MCP-1 overexpressing mouse model has shown an increased microglia accumulation around Aβ deposits and that this accumulation was accompanied by increased amount of diffuse plaques compared to single-transgenic Tg2576 littermates (Yamamoto, M., et al. (2005) Am. J Pathol. 166, 1475-1485).
MCP-1 levels are increased in CSF of AD patients and patients showing mild cognitive impairment (MCI) (Galimberti, D., et al., (2006) Arch. Neurol. 63, 538-543). Furthermore, MCP-1 shows an increased level in serum of patients with MCI and early AD (Clerici, F., et al., (2006) Neurobiol. Aging 27, 1763-1768).